Re: [Gluster-devel] afr, striping, and nufa scheduler

[Krishna Srinivas]

On 9/26/07, August R. Wohlt <address@hidden> wrote:
> So, if the cluster/afr volume is in the server specification instead of the
> client specification, and the 1st server in the afr array goes offline, the
> whole afr volume is unavailable for reads?
>
> I ask, because I noticed that if I had afr in server.vol it would hang when
> the other one went offline (in a :2 setup), but if I moved it to the client
> specification, it always works, even when the other half was offline. I
> think this is because in the client specification, I always had the local
> server first.

Can you explain using an example? I could not get it :-(

Krishna

>
> Thanks!
> :august
>
> On 9/25/07, Krishna Srinivas <address@hidden> wrote:
> >
> > You are right.
> >
> > For example:
> >
> > volume afr-example
> >   type cluster/afr
> >   subvolumes server1 server2
> >   option replicate *:2
> > end-volume
> >
> > Now afr-example will do all read operations from server1
> >
> > Krishna
> >
> >
> > On 9/25/07, Nathan Allen Stratton <address@hidden> wrote:
> > > On Tue, 25 Sep 2007, Krishna Srinivas wrote:
> > >
> > > > Hi August,
> > > >
> > > > Actually the read scheduler is not yet done, it is on our roadmap to
> > > > implement it. Now, AFR does all the read operations from its first
> > > > child.
> > >
> > > So if site A and B are AFR and site A is listed first both sites A and B
> > > will try to do all reads from A?
> > >
> > > -Nathan
> > >
> > >
> > > _______________________________________________
> > > Gluster-devel mailing list
> > > address@hidden
> > > http://lists.nongnu.org/mailman/listinfo/gluster-devel
> > >
> >
> >
> > _______________________________________________
> > Gluster-devel mailing list
> > address@hidden
> > http://lists.nongnu.org/mailman/listinfo/gluster-devel
> >
> _______________________________________________
> Gluster-devel mailing list
> address@hidden
> http://lists.nongnu.org/mailman/listinfo/gluster-devel
>